Poster
Sinkhorn Barycenters with Free Support via Frank-Wolfe Algorithm
Giulia Luise · Saverio Salzo · Massimiliano Pontil · Carlo Ciliberto
East Exhibition Hall B, C #113
Keywords: [ Theory ] [ Spaces of Functions and Kernels ] [ Convex Optimization; Theory ] [ Optimization ]
We present a novel algorithm to estimate the barycenter of arbitrary probability distributions with respect to the Sinkhorn divergence. Based on a Frank-Wolfe optimization strategy, our approach proceeds by populating the support of the barycenter incrementally, without requiring any pre-allocation. We consider discrete as well as continuous distributions, proving convergence rates of the proposed algorithm in both settings. Key elements of our analysis are a new result showing that the Sinkhorn divergence on compact domains has Lipschitz continuous gradient with respect to the Total Variation and a characterization of the sample complexity of Sinkhorn potentials. Experiments validate the effectiveness of our method in practice.
Live content is unavailable. Log in and register to view live content